asthma in pittsburgh and allegheny county, current information and
TRANSCRIPT
Asthma in Pittsburgh and
Allegheny County, Current
Information and Future
Directions
LuAnn Brink, Ph.D., M.P.H.
Visiting Assistant Professor
University of Pittsburgh Graduate School of Public Health
Department of Epidemiology
Objectives
• The increasing prevalence of asthma
• The local importance of asthma
• The history of asthma/air pollution
epidemiology
• The impact of air pollution on asthma
• The local impact of air pollution on
asthma
The burden of asthma
• In 2009, 25 million people, or 8% of the
population, had asthma.
– Compared to 2001, when 20 million, or 7%
– Asthma rates rose the most among black
children, an almost 50% increase
CDC Vital Signs “Asthma in the US,” May 2011
Asthma in Allegheny County
Adults by Race, 2002 and 2009-
2010
0
5
10
15
20
25
2002 2009-2010
Black
White
Adapted from AC BRFSS, published 2012
Adults with asthma in the US, 2009
SOURCE: Behavioral Risk Factor Surveillance System,
2009
Comparison of National and
Local Asthma Rates by Race,
2010
0
2
4
6
8
10
12
14
16
18
20
US PA AC
Current Asthma -- all
Current Asthma --
White
Current Asthma --
Black
Criteria Pollutants
• Ozone
• Nitrogen Oxides
• Sulfur Oxides
• Particulate Matter
• Carbon Monoxide
• Lead
John Balmes
Potential biological mechanism of PM
effects
Mechanisms of Ozone Toxicity
Direct oxidation
Free radical formation
Lipid peroxidation
Secondary inflammation/repair
Nitrogen Oxide
• NO2 not as potent of an oxidant as
ozone
• NO2 reacts with H2O to form HNO3
Sulfur Dioxide
• The NAAQS for SO2 allows for relatively
high short-term peak exposures.
• People with asthma are not protected
from exacerbations caused by brief
exposures.
Particulate Matter (PM)
• Several studies have documented increased respiratory symptoms or increased hospitalizations for acute respiratory illness in people in association with PM exposures.
• Decreased peak flow has been observed in panels of normal and asthmatic children in association with PM10.
The potential impact of traffic
pollution
• Several studies have shown increased
respiratory symptoms in children living
near roadways with increased traffic
density.
• Several studies have shown increased
asthma prevalence in relation to traffic
exposure (with NO2 often showing the
best single pollutant correlations).
Stebbings 1978
• Pulmonary Function Tests on 224 school
children during and after the Pittsburgh
air pollution episode of 11/75
• 4 exposed and 2 control schools
• Noted strong upward trends in Forced
Vital Capacity after episode
Delfino, 1994
• Hospital admissions for respiratory illnesses in Montreal between 1984-88 were 21.8% (9.7-33.8%) higher for 8-hour maximum increase of 38 ppb ozone in the summer – Among those >64 years of age
• Asthma admissions in May-October increased by 2.7% over mean levels for each 12 ug/m3 increase in PM10 levels 3 days prior to admission
• In July and August, admissions were 9.6% higher when SO4 had exceeded 8.1 ug/m3 4 days prior to admission day.
• PM10 had not exceeded the NAAQS of 150 ug/m3 during the time period.
Villeneuve – 2007
• Case-crossover study of asthma ED
visits and found that in the summer,
SO2, NO2, CO, PM2.5, PM10 and
ozone had significant effects .
– Children aged 2-4 and elderly were most
affected
• CO OR=1.48
• NO2 OR=1.5
Yap CA 2013
• Hospital admission in CA by zip code between 2000-2005
• Daily counts of respiratory admissions for – High and low SES by county
• Time series adjusting for time trends, seasonality, day of week, temperature, with pollution lags 0-6 days found an association of asthma and pollution
• For LA, Riverside, San Bernardino, and San Diego Counties, RR 1.03-1.07/10 ug PM2.5
July 27, 2011
The Relationship of Ambient Ozone and PM2.5
Levels and Asthma Emergency Department
Visits: Possible Influence of Gender and
Ethnicity Glad, Brink, Talbott, Lee, Xu, Saul, Rager
• Data from UPMC, which serves 60% of
Allegheny County
– 6979 patients seen in 6 EDs between
January 2002 and December 2005
– Discharged with asthma (ICD-9, 493.x)
Daily Air Pollution Data for AC
• Ozone and PM2.5 were obtained from
the ACHD Air Quality Program
– 3 ozone monitors
– 2 continuously operating PM2.5 monitors
• Used ones in center of city, near hospitals of
interest
– Daily 1-hour maximum ozone
– Daily mean PM2.5
– Daily mean temperature and humidity
Study Population
• 60.3% Caucasian
– Mean age 42.4
– 17.7% over age 65
– 10.9% under age 14
• 37.8% African Americans
– Mean age 35.2
– 7.6% over age 65
– 18.9% under age 14
Visits for Asthma
• Ranged from 1-19 visits per person
• African Americans had slightly more
visits within the time period
The Case-Crossover Design
• First proposed by Maclure in 1991
• Originally designed to avoid selection bias from a case-control study
• Designed to answer the question “Is a particular health event triggered by something that happened just before the health event”
• Basic idea: compare a patient’s exposure experience on the day of their outcome (heart attack) with their exposure experience on the day before
Features of Case-Crossover Design
• Only cases are analyzed
• The same individuals “cross over”
between being cases and being
controls
– The idea is that people cross over
between short periods of exposure to
hypothetical triggers and much longer
periods of unexposed time.
Selecting the Referent (Control) Times
• Referent times act as the individually-matched “controls”
• Approaches to selecting a referent period – Basic approach - Match one hazard period
to one referent period (matched pair interval approach)
– Multiple interval approach – Match one hazard period to multiple referent periods • Symmetric bi-directional approach
• Time-stratified approach
Sun Mon Tue Wed Thu Fri Sat
1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30
49
Time Stratified Sampling Approach
Confounding
• The case-crossover design controls for measured and unmeasured confounders that do not change over time, such as age, gender, genes.
• Confounding by time-varying factors is possible – e.g., seasonal patterns, day of week, long term trends – Can be limited by choice of referent
time periods
Glad, Brink, Talbott 2012
Conclusions
• A 10 ppb increase in the 1-hour daily maximum ozone level was significantly related to a 2.5% increase in asthma ED visits 2 days later
• When considering PM2.5 also, ozone contributed a 2.1% increase 2 days later.
• One day after PM2.5 exposure, a 3.6 % increase in asthma ED visits occurred
Conclusions continued
• Although the entire population were
affected 2 days after a PM2.5 pollution
event, this effect was not significant
among Caucasians (1.015, 0.989-1.041)
– It was higher among African Americans,
1.025, 0.997-1.053, driving the overall
effect
Possible reasons for differences
• Access to medical care
• Access to air conditioning
• Intrinsic differences
• Different distributions of pollutants
Strengths and Limitations
• ED visits provide a strong and specific outcome measure
• Use of case-crossover design allows control for seasonality, secular trends, and time-invariant factors
• Use of a single monitor to estimate exposure
• Sample of ED visits may not be representative of all ED visits
Current Work:
Asthma Predictive
Modeling
Sharma, Brink
Statistical Methodology
• Mixed Model longitudinal analysis
using Poisson regression is used analyze
daily times series of asthma, circulatory
and respiratory counts from 2004-2005
• Risks are estimated and presented in
the tables as rate ratio
Results to date
• Based upon asthma hospitalization
occurring in 2004-2005, a significant
increase in asthma hospitalizations with
same-day increase in PM2.5 was
noted.
Conclusions
• Recent studies conducted in Pittsburgh
indicate an effect of both ozone and
PM2.5 on asthma exacerbations in
Pittsburgh, PA